ABSTRACT
Background/Aims: Digestive symptoms are common in patients with COVID-19. Neverthe-less, the evidence available so far is based on retrospective and observational studies. This prospective multicenter cohort study aimed to describe the frequency, intensity, evolution, and impact of digestive symptoms and complications, during hospitalization and after dis-charge, of patients with COVID-19. Methods: Patients hospitalized due to COVID-19 (posi-tive PCR for SARS-CoV-2) from May to August 2020, were prospectively recruited in 31 centers. Follow-up included the period between admission and 15 days after discharge. Results: 829 patients (mean age 56.7±17.9 years;42% of females) were enrolled in this study. Of these, 7.2% were active smokers and the mean BMI was 29.1±5.7. Proton pump inhibitors were used by 21.5% (n=178). The most prevalent symptoms on admission were diarrhea (39.4%), nausea (27.4%), and abdominal pain (20.7%). Anorexia, a non-specific symptom, was present in 49.8% of hospitalized patients. At discharge and 15 days after discharge, most symptoms resolved, returning to the baseline prevalence of patients (<5%). Digestive complications during admission were infrequent, except for liver injury defined as hypertransaminasemia which was present in 267 patients (32.3%). The mean length of hospital stay was 8 days (5-12) and 13.6% needed ICU admission. Death happened in 5.2%of patients. On multivariate analysis, diarrhea on admission was associated with a shorter hospital stay (<10 days) ORa 0.508 (0.350-0.739) p=0.000. During hospitalization, diarrhea, constipation, and abdominal bloating were associated with shorter hospital stay ORa 0.531 (0.298-0.946) p=0.032, ORa 0.384 (0.167-0.885) p=0.025, ORa 0.163 (0.057-0.466) p= 0.00, respectively.Odynophagia and dysphagia during hospitalization were associated with a higher need for ICU admission, ORa 6.518 (2.255-18.835) p=0.001 and ORa 4.035 (1.453-11.204) p=0.007, respectively. Liver injury during hospitalization was associated with a higher hospital stay (>10 days) ORa 1.442 (1.019-2.041) p=0.039. In the linear regression analysis, the set of GI symptoms and complications, along with age, comorbidity, and respiratory symptoms, were able to predict 43% (R2 0.43) of the observed variability in the speed of ICU admission;in this case, digestive symptoms slowed it down (more days until ICU admission). Conclusions: Gastrointestinal manifestations of COVID-19 are common in hospitalized patients, while complications are infrequent. Gastrointestinal symptoms seemed to predict a shorter hospital stay and slower speed of ICU admission. These tend to resolve to their baseline prevalence 15 days after discharge, while elevated transaminases were associated with a longer hospital stay. Odynophagia and dysphagia during hospitaliza-tion were associated with an increased need for ICU admission.